Connections among Global Mortality, Health Care Coverage, Education, Economy, and Pollution

DATA 604 (Fall 2020) - Group 6: Giesbrecht, Lam, Leung, Nelson, Nguyen

#conda install -c plotly plotly

import mysql.connector
import pandas as pd
from mysql.connector import errorcode
import plotly.express as px

Introduction

Purpose: Study the relationship between global data of stressors in countries that could indicate a population’s prosperity and death by popular diseases

Examples of relationships:

• Mortality: increases in poverty, poor living environment, lack of education, & health care policies
• Education: facilitates economic growth, awareness of environmental sustainability, effective policies
• Economic activities: can lead to environmental issues, health quality
• Pollution: and Health coverage: can affect health/mortality severely

Studying the potential relationships could help a country to better understand how to better deal with popular deadly diseases and in what areas they could improve to lower the death counts

Datasets

• Annual Number of Deaths By Cause (Our World in Data, 2019)
  
• Average Years of Education (Our World in Data, 2018)
  
• National GDP Per Capita (UNESCO Institute for Statistics, 2020)
  
• Air Pollution (World Bank database, 2020)
  
• Universal Health Coverage (GHDx, 2019)
  

Data Cleaning and Wrangling

• Pollution dataset needed to be reorganized to include years as one column instead of each year as their own column
• Universal Health Coverage needed a column for country code
• All datasets were contructed with Country_Year, Country_Code, Year as columns and additional columns for the associated topic value.
• Inner join

# Read in both data files into pandas for use in plotly 
#data = pd.read_csv('combined_data.csv')
#data1 = pd.read_csv('uhc_data.csv')

• Comparability: datasets have common attributes that can be used as keys
  • Year
  • Country_Code

• Converted raw data into a ranking WHERE a rank of 1 would be the highest value of the category being looked at.

Mean annual pollution exposure rank, mean school year rank, GDP per capita rank, UHC index rank for year 1990

#open connection to execute an SQL query
try:
    conn = mysql.connector.connect(option_files=['connection.conf','password.conf'])
    db_cursor = conn.cursor()
    
    #enter query here
    query1 = ('''
 SELECT 
	c.`Year`, 
    c.Country_name, 
    c.Mean_Annual_Pollution_Exposure_Rank, 
    c.Mean_School_Year_Rank, 
    c.GDP_Per_Capita_rank, 
    rank() over (partition by (u.year) order by u.UHC_Index desc) as UHC_Index_rank
FROM `l02-6`.combined AS c 
	join 
		(select Year, Country_Code, UHC_Index 
         from `l02-6`.uhc) as u 
	on c.Country_code=u.Country_code and c.Year = u.Year
order by c.Mean_Annual_Pollution_Exposure_rank asc;''')
    
    #copy query result to the new list
    data1 = pd.read_sql(query1, conn)
    
    db_cursor.close()

except mysql.connector.Error as err:
  if err.errno == errorcode.ER_ACCESS_DENIED_ERROR:
    print("Something is wrong with your user name or password")
  elif err.errno == errorcode.ER_BAD_DB_ERROR:
    print("Database does not exist")
  else:
    print(err)
else:
  conn.close()

display(data1.head())
display(data1.tail())

	Year	Country_name	Mean_Annual_Pollution_Exposure_Rank	Mean_School_Year_Rank	GDP_Per_Capita_rank	UHC_Index_rank
0	1990	Qatar	1	73	20	29
1	1990	Niger	2	128	105	127
2	1990	Nepal	3	120	126	110
3	1990	India	4	108	112	120
4	1990	Egypt	5	99	91	94

	Year	Country_name	Mean_Annual_Pollution_Exposure_Rank	Mean_School_Year_Rank	GDP_Per_Capita_rank	UHC_Index_rank
124	1990	Canada	125	11	14	2
125	1990	Sweden	126	8	3	3
126	1990	Finland	127	47	4	9
127	1990	New Zealand	128	2	23	13
128	1990	Brunei	129	46	24	55

Total deaths by disease rank, mean annual pollution exposure rank, mean school year rank,GDP per capita rank over the years 1990 - 2017

#open connection to execute an SQL query
try:
    conn = mysql.connector.connect(option_files=['connection.conf','password.conf'])
    db_cursor = conn.cursor()
    
    #enter query here
    query3 = ('''
SELECT 
pd.Country_Name, 
pd.Country_Code,
pd.Year,
pd.Liver_disease,
pd.Digestive_Disease,
pd.HIV_AIDS,
pd.Diarrheal_diseases,
pd.Intestinal_Infectious_Disease,
pd.Lower_Respiratory_Infections,
pd.Meningitis,
pd.Alzheimer_disease_and_other_dementias,
pd.Parkinson_disease,
pd.Malaria,
pd.Maternal_disorders,
pd.Neonatal_disorders,
pd.Nutritional_deficiencies,
pd.Diabetes_mellitus,
pd.Chronic_kidney_disease,
pd.Chronic_respiratory_diseases,
pd.Neoplasms,
pd.Protein_Energy_Malnutrition,
pd.Cardiovascular_Diseases,
pd.Mean_Annual_Pollution_Exposure,
eg.Mean_School_Year,
eg.GDP_Per_Capita,
cume_dist() over (partition by (year) order by pd.liver_disease ) * 100 as Liver_disease_rank,
cume_dist() over (partition by (year) order by pd.Digestive_Disease ) * 100 as Digestive_disease_rank,
cume_dist() over (partition by (year) order by pd.HIV_AIDS ) * 100 as HIV_AIDS_rank,
cume_dist() over (partition by (year) order by pd.Diarrheal_diseases ) * 100 as Diarrheal_disease_rank,
cume_dist() over (partition by (year) order by pd.Intestinal_Infectious_Disease ) * 100 as Intestinal_Infectious_disease_rank,
cume_dist() over (partition by (year) order by pd.Lower_Respiratory_Infections ) * 100 as Lower_Respiratory_Infections_rank,
cume_dist() over (partition by (year) order by pd.Meningitis ) * 100 as Meningitis_rank,
cume_dist() over (partition by (year) order by pd.Alzheimer_disease_and_other_dementias ) * 100 as Alzheimer_disease_and_other_dementias_rank,
cume_dist() over (partition by (year) order by pd.Parkinson_disease ) * 100 as Parkinson_disease_rank,
cume_dist() over (partition by (year) order by pd.Malaria ) * 100 as Malaria_rank,
cume_dist() over (partition by (year) order by pd.Maternal_disorders ) * 100 as Maternal_disorders_rank,
cume_dist() over (partition by (year) order by pd.Neonatal_disorders ) * 100 as Neonatal_disorders_rank,
cume_dist() over (partition by (year) order by pd.Nutritional_deficiencies ) * 100 as Nutritional_deficiencies_rank,
cume_dist() over (partition by (year) order by pd.Diabetes_mellitus ) * 100 as Diabetes_mellitus_rank,
cume_dist() over (partition by (year) order by pd.Chronic_kidney_disease ) * 100 as Chronic_kidney_disease_rank,
cume_dist() over (partition by (year) order by pd.Chronic_respiratory_diseases ) * 100 as Chronic_respiratory_diseases_rank,
cume_dist() over (partition by (year) order by pd.Neoplasms ) * 100 as Neoplasms_rank,
cume_dist() over (partition by (year) order by pd.Protein_Energy_Malnutrition ) * 100 as Protein_Energy_Malnutrition_rank,
cume_dist() over (partition by (year) order by pd.Cardiovascular_Diseases ) * 100 as Cardiovascular_Diseases_rank,
cume_dist() over (partition by (year) order by pd.Mean_Annual_Pollution_Exposure ) * 100 as Mean_Annual_Pollution_Exposure_rank,
cume_dist() over (partition by (year) order by eg.Mean_School_Year ) * 100 as Mean_School_Year_rank,
cume_dist() over (partition by (year) order by eg.GDP_Per_Capita ) * 100 as GDP_Per_Capita_rank

FROM

(SELECT d.*, 
p.Mean_Annual_Pollution_Exposure
FROM `l02-6`.death_type AS d JOIN `l02-6`.pollution AS p 
ON d.Country_Code = p.Country_Code AND d.Year = p.Year) AS pd

JOIN 

(SELECT e.*, g.GDP_Per_Capita
FROM `l02-6`.avg_education AS e JOIN `l02-6`.gdp_data AS g 
ON e.Country_Code = g.Country_Code AND e.Year = g.Year) AS eg 

ON pd.Country_Code = eg.Country_Code AND pd.Year = eg.Year
order by Year, Mean_Annual_Pollution_Exposure_rank;
''')
    
    #copy query result to the new list
    data3 = pd.read_sql(query3, conn)
    
    db_cursor.close()

except mysql.connector.Error as err:
  if err.errno == errorcode.ER_ACCESS_DENIED_ERROR:
    print("Something is wrong with your user name or password")
  elif err.errno == errorcode.ER_BAD_DB_ERROR:
    print("Database does not exist")
  else:
    print(err)
else:
  conn.close()

display(data3.head())
display(data3.tail())

	Country_Name	Country_Code	Year	Liver_Disease	Digestive_Disease	HIV_AIDS	Diarrheal_diseases	Intestinal_Infectious_Disease	Lower_Respiratory_Infections	Meningitis	...	Nutritional_deficiencies_rank	Diabetes_mellitus_rank	Chronic_kidney_disease_rank	Chronic_respiratory_diseases_rank	Neoplasms_rank	Protein_Energy_Malnutrition_rank	Cardiovascular_Diseases_rank	Mean_Annual_Pollution_Exposure_rank	Mean_School_Year_rank	GDP_Per_Capita_rank
0	Brunei	BRN	1990	14	32	1	1	0	38	4	...	3.100775	8.527132	3.100775	5.426357	3.875969	6.201550	2.325581	0.775194	65.116279	82.170543
1	New Zealand	NZL	1990	195	764	57	13	0	875	28	...	13.178295	26.356589	24.806202	39.534884	51.937984	15.503876	44.186047	1.550388	99.224806	82.945736
2	Finland	FIN	1990	572	1687	16	58	0	2007	44	...	11.627907	34.108527	19.379845	33.333333	61.240310	9.302326	57.364341	2.325581	64.341085	97.674419
3	Sweden	SWE	1990	827	2859	70	42	1	3702	71	...	25.581395	56.589147	37.984496	55.813953	72.093023	28.682171	72.868217	3.100775	94.573643	98.449612
4	Canada	CAN	1990	2944	6928	797	54	2	5690	153	...	44.961240	80.620155	73.643411	78.294574	86.821705	44.186047	80.620155	3.875969	92.248062	89.922481

5 rows × 47 columns

	Country_Name	Country_Code	Year	Liver_Disease	Digestive_Disease	HIV_AIDS	Diarrheal_diseases	Intestinal_Infectious_Disease	Lower_Respiratory_Infections	Meningitis	...	Nutritional_deficiencies_rank	Diabetes_mellitus_rank	Chronic_kidney_disease_rank	Chronic_respiratory_diseases_rank	Neoplasms_rank	Protein_Energy_Malnutrition_rank	Cardiovascular_Diseases_rank	Mean_Annual_Pollution_Exposure_rank	Mean_School_Year_rank	GDP_Per_Capita_rank
1717	Saudi Arabia	SAU	2017	2461	3109	526	212	0	3505	145	...	40.555556	58.888889	75.000000	56.666667	58.333333	42.777778	65.555556	97.777778	57.222222	80.555556
1718	India	IND	2017	217896	419545	52106	719083	0	507364	50052	...	100.000000	100.000000	100.000000	100.000000	99.444444	99.444444	99.444444	98.333333	27.222222	25.555556
1719	Qatar	QAT	2017	114	159	7	3	0	91	6	...	5.555556	22.777778	13.888889	6.666667	15.000000	3.333333	12.777778	98.888889	61.111111	97.222222
1720	Niger	NER	2017	1504	4412	1951	21955	0	19710	7815	...	84.444444	52.777778	56.666667	58.888889	47.777778	84.444444	45.555556	99.444444	1.111111	3.888889
1721	Nepal	NPL	2017	5671	9756	2619	10796	0	9297	837	...	74.444444	78.333333	81.111111	90.000000	68.888889	69.444444	73.888889	100.000000	16.111111	12.777778

5 rows × 47 columns

Global Death, GDP Per Capita, Pollution, Education, and Health Coverage in 1990

• Combine all five datasets (INNER JOIN)
• Add a rank column for each field/variable, by year using RANK() function (1st rank = highest value of the year)

Or:

• Combined Table + Health Coverage Table
• Rank each country's Health Coverage

#Combining all five datasets


#open connection to execute an SQL query
try:
    conn = mysql.connector.connect(option_files=['connection.conf','password.conf'])
    db_cursor = conn.cursor()
    
    #enter query here
    query4 = ('''SELECT 
                    c.*,
                    u.UHC_Index,
                    RANK() OVER (PARTITION BY Year ORDER BY u.UHC_Index DESC) AS UHC_Index_rank
                FROM
                    combined AS c JOIN uhc AS u USING(Country_Code, Year);
            ''')
    
    #get query results
    data4 = pd.read_sql(query4, conn)
    
    db_cursor.close()

except mysql.connector.Error as err:
  if err.errno == errorcode.ER_ACCESS_DENIED_ERROR:
    print("Something is wrong with your user name or password")
  elif err.errno == errorcode.ER_BAD_DB_ERROR:
    print("Database does not exist")
  else:
    print(err)
else:
  conn.close()

display(data4.head())
display(data4.tail())

	Country_Name	Country_code	Year	Liver_Disease	Digestive_Disease	HIV_AIDS	Diarrheal_diseases	Intestinal_Infectious_Disease	Lower_Respiratory_Infections	Meningitis	...	Chronic_kidney_disease_rank	Chronic_respiratory_diseases_rank	Neoplasms_rank	Protein_Energy_Malnutrition_rank	Cardiovascular_Diseases_rank	Mean_Annual_Pollution_Exposure_rank	Mean_School_Year_rank	GDP_Per_Capita_rank	UHC_Index	UHC_Index_rank
0	Japan	JPN	1990	22684	38521	84	951	6	54123	734	...	5	12	5	62	8	107	15	9	80.8817	1
1	Canada	CAN	1990	2944	6928	797	54	2	5690	153	...	35	29	18	73	26	125	11	14	79.7969	2
2	Sweden	SWE	1990	827	2859	70	42	1	3702	71	...	81	58	37	93	36	126	8	3	78.7949	3
3	Iceland	ISL	1990	12	46	2	1	0	83	3	...	129	125	119	127	123	124	30	8	78.2034	4
4	Switzerland	CHE	1990	806	1732	176	75	0	2250	143	...	92	59	42	84	53	108	12	1	77.1674	5

5 rows × 49 columns

	Country_Name	Country_code	Year	Liver_Disease	Digestive_Disease	HIV_AIDS	Diarrheal_diseases	Intestinal_Infectious_Disease	Lower_Respiratory_Infections	Meningitis	...	Chronic_kidney_disease_rank	Chronic_respiratory_diseases_rank	Neoplasms_rank	Protein_Energy_Malnutrition_rank	Cardiovascular_Diseases_rank	Mean_Annual_Pollution_Exposure_rank	Mean_School_Year_rank	GDP_Per_Capita_rank	UHC_Index	UHC_Index_rank
124	Democratic Republic of Congo	COD	1990	6771	12854	19707	34416	834	52954	9414	...	22	28	39	4	37	23	115	120	22.9475	125
125	Mali	MLI	1990	1826	3405	502	23295	641	12200	6045	...	47	46	66	7	69	26	127	118	20.9199	126
126	Niger	NER	1990	1068	2675	199	32061	663	25299	12367	...	55	67	79	23	88	2	128	105	17.9624	127
127	Laos	LAO	1990	877	1722	0	4367	431	10014	940	...	69	61	88	44	72	60	105	124	15.4171	128
128	Haiti	HTI	1990	1004	2433	2340	10469	7	7937	1743	...	62	64	71	32	60	93	111	106	11.8838	129

5 rows × 49 columns

A closer look

"Combined" dataset: A result of Albert's second query. Contains the ranking of deaths by diseases, pollution, education, and gdp per capita.

First Query:

• Do an "as close as possible" join between the combined and UHC datasets for the year 2010.

Second Query:

• Each country's average rank in their first and last recorded years.

#open connection to execute an SQL query
try:
    conn = mysql.connector.connect(option_files=['connection.conf','password.conf'])
    
    # Query to do an approximate join between the table that contains everything except the UHC
    # dataset, and the UHC dataset for the year 2010.
    query5 = ('''
    select
        u.Year as corresponding_uhc_year,
        u.uhc_index_rank,
        comb.*
    from
        (select c.* from combined c join
            (select Country_Code, Country_Name, group_concat(Year order by abs(Year-2010)) nearest2010 from
                combined group by Country_Name) t2
            on c.Country_Name=t2.Country_Name and c.Year=t2.nearest2010
        ) comb,
    (select
        Country_Code, Country_Name, Year,
            ntile(200) over (partition by Year order by uhc_index) as uhc_index_rank
    from uhc where year=2010) u
    where
        comb.Country_Code=u.Country_Code;
    ''')
    
    # Query to find the average of pollution, education, and gdp for each country in the first 
    # and last year that they are present in the dataset that contains everything except 
    # the UHC dataset
    query6 = ('''
    select 
        first.Country_Name, first.Year firstYear,
        ((184-first.mean_annual_pollution_exposure_rank) + first.mean_school_year_rank +
            first.gdp_per_capita_rank)/3 first_avg,
        last.Year lastYear,
        ((184-last.mean_annual_pollution_exposure_rank) + last.mean_school_year_rank +
            last.gdp_per_capita_rank)/3 last_avg
    from 
        (select c.* from combined c join
            (select Country_Name, min(Year) minyear from combined group by Country_Name) t2
            on c.Country_Name = t2.Country_Name and c.Year = t2.minyear
        ) first,
        (select c.* from combined c join
            (select Country_Name, max(Year) maxyear from combined group by Country_Name) t2
            on c.Country_Name = t2.Country_Name and c.Year = t2.maxyear
        ) last
    where 
        first.Country_Code = last.Country_Code order by first.Country_Code;
    ''')
    
    data5 = pd.read_sql(query5, conn)
    data6 = pd.read_sql(query6, conn)
       
    
except mysql.connector.Error as err:
  if err.errno == errorcode.ER_ACCESS_DENIED_ERROR:
    print("Something is wrong with your user name or password")
  elif err.errno == errorcode.ER_BAD_DB_ERROR:
    print("Database does not exist")
  else:
    print(err)
else:
  conn.close()

# Results of the first query.
display(data5.set_index("Country_Name").head())
display(data5.set_index("Country_Name").tail())

	corresponding_uhc_year	uhc_index_rank	Country_code	Year	Liver_Disease	Digestive_Disease	HIV_AIDS	Diarrheal_diseases	Intestinal_Infectious_Disease	Lower_Respiratory_Infections	...	Nutritional_deficiencies_rank	Diabetes_mellitus_rank	Chronic_kidney_disease_rank	Chronic_respiratory_diseases_rank	Neoplasms_rank	Protein_Energy_Malnutrition_rank	Cardiovascular_Diseases_rank	Mean_Annual_Pollution_Exposure_rank	Mean_School_Year_rank	GDP_Per_Capita_rank
Country_Name
Afghanistan	2010	4	AFG	2011	1989	4850	177	7053	300	23115	...	38	38	43	41	60	38	43	10	166	175
Angola	2010	6	AGO	2011	4451	7113	13797	21595	415	18432	...	11	53	59	67	77	10	82	51	151	98
Azerbaijan	2010	41	AZE	2011	2775	3492	37	238	0	3516	...	115	89	94	89	82	111	55	102	50	78
Bangladesh	2010	84	BGD	2011	25375	41026	442	26979	13008	52920	...	8	8	13	5	21	7	11	8	144	160
Belize	2010	91	BLZ	2011	41	69	73	11	0	93	...	139	163	165	171	170	134	171	74	54	101

5 rows × 48 columns

	corresponding_uhc_year	uhc_index_rank	Country_code	Year	Liver_Disease	Digestive_Disease	HIV_AIDS	Diarrheal_diseases	Intestinal_Infectious_Disease	Lower_Respiratory_Infections	...	Nutritional_deficiencies_rank	Diabetes_mellitus_rank	Chronic_kidney_disease_rank	Chronic_respiratory_diseases_rank	Neoplasms_rank	Protein_Energy_Malnutrition_rank	Cardiovascular_Diseases_rank	Mean_Annual_Pollution_Exposure_rank	Mean_School_Year_rank	GDP_Per_Capita_rank
Country_Name
Norway	2010	195	NOR	2011	368	1389	21	342	0	1859	...	114	121	113	76	73	113	95	176	5	2
Singapore	2010	194	SGP	2011	213	459	20	6	0	2050	...	173	147	127	129	111	166	127	128	33	9
Spain	2010	191	ESP	2011	7884	19941	961	813	3	11208	...	69	27	20	14	13	80	23	169	71	29
Sweden	2010	190	SWE	2011	1026	2943	33	612	0	2734	...	100	78	90	56	45	98	56	181	13	7
Switzerland	2010	197	CHE	2011	917	2516	42	183	0	1447	...	108	101	85	78	57	110	78	164	2	3

5 rows × 48 columns

# Results of the second query.
data6.sort_values(by="last_avg", axis=0, inplace=True)
# Get the top 10 and bottom 10.
data_top = data6.iloc[:10]
data_bottom = data6.iloc[-10:]
# Melt the dataframes, for plotting purposes.
df_top = pd.melt(data_top.loc[:, ["Country_Name", "first_avg", "last_avg"]],
                id_vars=["Country_Name"], var_name="Year",
                value_name="value")
df_bottom = pd.melt(data_bottom.loc[:, ["Country_Name", "first_avg", "last_avg"]],
                id_vars=["Country_Name"], var_name="Year",
                value_name="value")
# Sort them, again for plotting purposes.
df_top.sort_values(by=["Year", "value"], ascending=[True, True], axis=0, inplace=True)
df_bottom.sort_values(by=["Year", "value"], ascending=[True, True], axis=0, inplace=True)

fig_top = px.bar(df_top, x="Country_Name", y="value", color="Year", barmode="group",
                title="Top 10 Countries In Their First Year",
                labels={"Country_Name": "Country Name",
                       "value":"Average Rank (out of 183)"})


fig_bottom = px.bar(df_bottom, x="Country_Name", y="value", color="Year", barmode="group",
                title="Bottom 10 Countries In Their First Year",
                labels={"Country_Name": "Country Name",
                       "value":"Average Rank (out of 183)"})

fig_top.show()

fig_bottom.show()

UHC Index and Death Count

• Computing ranks across UHC index and total death count for the years 1990 and 2010 for all countries

• How do the two variables relate to each other?

#open connection to execute an SQL query
try:
    conn = mysql.connector.connect(option_files=['connection.conf','password.conf'])
    db_cursor = conn.cursor()
    
    #enter query here
    
    query7 = ('''SELECT
                    u.Country_Name, 
                    u.Country_Code,
                    u.Year,
                    u.UHC_Index,
                    (pd.Liver_disease +
                    pd.Digestive_Disease +
                    pd.HIV_AIDS +
                    pd.Diarrheal_diseases +
                    pd.Intestinal_Infectious_Disease +
                    pd.Lower_Respiratory_Infections +
                    pd.Meningitis +
                    pd.Alzheimer_disease_and_other_dementias +
                    pd.Parkinson_disease +
                    pd.Malaria +
                    pd.Maternal_disorders +
                    pd.Neonatal_disorders +
                    pd.Nutritional_deficiencies +
                    pd.Diabetes_mellitus +
                    pd.Chronic_kidney_disease +
                    pd.Chronic_respiratory_diseases +
                    pd.Neoplasms +
                    pd.Protein_Energy_Malnutrition +
                    pd.Cardiovascular_Diseases) AS Total_Death,
                    RANK() OVER (PARTITION BY Year ORDER BY u.UHC_Index DESC) AS UHC_Index_Rank,
                    RANK() OVER (PARTITION BY Year ORDER BY Total_Death DESC) AS Total_Death_Rank
                FROM
                    death_type AS pd JOIN uhc AS u USING(Country_Code, Year);''')
            
    
    #copy query result to the new list
    data7 = pd.read_sql(query7, conn)
    
    db_cursor.close()

except mysql.connector.Error as err:
  if err.errno == errorcode.ER_ACCESS_DENIED_ERROR:
    print("Something is wrong with your user name or password")
  elif err.errno == errorcode.ER_BAD_DB_ERROR:
    print("Database does not exist")
  else:
    print(err)
else:
  conn.close()

display(data7.head())
display(data7.tail())

	Country_Name	Country_Code	Year	UHC_Index	Total_Death	UHC_Index_Rank	Total_Death_Rank
0	China	CHN	1990	49.8103	6793986	82	1
1	India	IND	1990	26.9491	6368008	170	2
2	United States of America	USA	1990	73.4118	1942091	7	3
3	Russian Federation	RUS	1990	54.239	1396952	57	4
4	Nigeria	NGA	1990	22.5911	1093384	180	5

	Country_Name	Country_Code	Year	UHC_Index	Total_Death	UHC_Index_Rank	Total_Death_Rank
383	Bermuda	BMU	2010	73.7297	414	37	190
384	Marshall Islands	MHL	2010	42.1253	326	144	191
385	Greenland	GRL	2010	63.5621	324	59	192
386	American Samoa	ASM	2010	53.9934	254	100	193
387	Northern Mariana Islands	MNP	2010	63.6531	191	58	194

Visualizing relationships from the combined dataset

• Meningitis (inflammation of the meninges that surround brain and spinal cord)
• Diabetes Mellitus (insulin vs glucose)
• Neoplasms (tumors both benign and malignant)

fig1 = px.scatter(data2,x='Mean_Annual_Pollution_Exposure_rank',y='Meningitis_rank',trendline = 'lowess')
fig1.data[1].update(line_color='red') 
fig1.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig2 = px.scatter(data2,x='GDP_Per_Capita_rank',y='Meningitis_rank',trendline = 'lowess')
fig2.data[1].update(line_color='red') 
fig2.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig3 = px.scatter(data2,x='Mean_School_Year_rank',y='Meningitis_rank',trendline = 'lowess')
fig3.data[1].update(line_color='red') 
fig3.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig1.show()

fig2.show()

fig3.show()

fig4 = px.scatter(data2,x='Mean_Annual_Pollution_Exposure_rank',y='Diabetes_mellitus_rank',trendline = 'lowess')
fig4.data[1].update(line_color='red') 
fig4.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig5 = px.scatter(data2,x='GDP_Per_Capita_rank',y='Diabetes_mellitus_rank',trendline = 'lowess')
fig5.data[1].update(line_color='red') 
fig5.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig6 = px.scatter(data2,x='Mean_School_Year_rank',y='Diabetes_mellitus_rank',trendline = 'lowess')
fig6.data[1].update(line_color='red') 
fig6.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig4.show()

fig5.show()

fig6.show()

fig7 = px.scatter(data2,x='Mean_Annual_Pollution_Exposure_rank',y='Neoplasms_rank',trendline = 'lowess')
fig7.data[1].update(line_color='red') 
fig7.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig8 = px.scatter(data2,x='GDP_Per_Capita_rank',y='Neoplasms_rank',trendline = 'lowess')
fig8.data[1].update(line_color='red') 
fig8.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig9 = px.scatter(data2,x='Mean_School_Year_rank',y='Neoplasms_rank',trendline = 'lowess')
fig9.data[1].update(line_color='red') 
fig9.update_layout(autosize=False,width=1000,height=500,margin
                   =dict(l=50,r=50,b=100,t=100,pad=4),paper_bgcolor="LightSteelBlue",)

fig7.show()

fig8.show()

fig9.show()

Where are we going next?

• Look at all relationships and categorize them into which diseases relate to which predictors to find out an overall picture

• Add more factors within countries that describe other areas of a countries economic, cultural or educational standing

• Population is a key factor not studied - Countries like India and China with large populations may end up skewing the data

  • Could have used deaths/100,000 people for a direct comparison or a potential ratio between deaths and predictor
• Additonal visualizations can be made to demonstrate patterns to yet be found in the data

Thank you!

References

Canadian Cancer Society, 2020. How Cancer Grows. [image] Available at: https://www.cancer.ca/en/cancer-information/cancer-101/what-is-cancer/how-cancer-starts-grows-and-spreads/?region=on [Accessed 2 December 2020].

Diabetes Library, 2020. How Does Insulin Work?. [image] Available at: https://diabeteslibrary.org/insulin-vs-glucagon/ [Accessed 2 December 2020].

Medical Stocks, 2020. Meninges Medical 3D Illustration. [image] Available at: https://www.dreamstime.com/meninges-medical-d-illustration-white-background-eps-cross-view-human-brain-meninges-medical-d-illustration-white-image142325632 [Accessed 2 December 2020].